Workpieces sorting system based on industrial robot of machine vision

Xia, Ke; Weng, Zhengxin

doi:10.1109/icsai.2016.7810992

Cited by 21 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, based on Hu moment invariance extraction, the contours of the workpieces are extracted and matched with the template to identify the target workpieces [7]. Obtaining the contours information of the workpieces through the Canny algorithm, combining the probabilistic Huff transform and Freeman chain code to identify the shape and calculate the center space coordinates of the workpieces [8]. Since such algorithms mostly use global search, they are timeconsuming [9], and are sensitive to factors such as shadows and noise [10,11].…”

Section: Related Workmentioning

confidence: 99%

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Lei

Yao

Chen

et al. 2020

2020 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

A workpiece detection method based on fusion of deep learning and image processing is proposed. Firstly, the workpiece bounding boxes are located in the workpiece images by YOLOv3, whose parameters are compressed by an improved convolutional neural network residual structure pruning strategy. Then, the workpiece images are cropped based on the bounding boxes with cropping biases. Finally, the contours and suitable gripping points of the workpieces are obtained through image processing. The experimental results show that mean Average Precision (mAP) is 98.60% for YOLOv3, and 99.38% for that one by pruning 50.89% of its parameters, and the inference time is shortened by 31.13%. Image processing effectively corrects the bounding boxes obtained by deep learning, and obtains workpiece contour and gripping point information.

show abstract

Section: Related Workmentioning

confidence: 99%

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Lei

Yao

Chen

et al. 2020

2020 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

show abstract

“…Machine Vision is one of the most frontiers and revolutionary technology in computer science as a plethora of industrial activities has been potentially benefitted. It ensures consistent and continuous excellence in automating monotonous chores-visual inspection in manufacturing [1][2][3], localization and navigation for robotic guidance [4], real-time measuring and sorting in factory floors and production lines [5][6][7]. Vision systems bring operational benefits by reducing human involvement in a manufacturing process and excels in quantitative analysis of structured scenes because of their speed, accuracy and repeatability.…”

Section: Introductionmentioning

confidence: 99%

Design and implementation of shape-based feature extraction engine for vision systems using Zynq SoC

Mohan

Kurian

2022

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

With the great impact of vision and Artificial Intelligence (AI) technology in the fields of quality control, robotic assembly and robot navigation, the hardware implementation of object detection and classification algorithms on embedded platforms has got ever-increasing attention these days. The real-time performance with optimum resource utilization of the implementation and its reliability as well as the robustness of the underlying algorithm is the overarching challenges in this field. In this work, an approach employing a fast and accurate vision-based shape-detection algorithm has been proposed and its implementation in heterogeneous System on Chip (SoC) is discussed. The proposed system determines centroid distance and its Fourier Transform for the object feature vector extraction and is realized in the Zybo Z7 development board. The ARM processor is responsible for communication with the external systems as well as for writing data to the Block RAM (BRAM), the control signals for efficient execution of the memory operations are designed and implemented using Finite State Machine (FSM) in the Programmable Logic (PL) fabric. Shape feature vector determination has been accelerated using custom modules developed in Verilog, taking full advantage of the possible parallelization and pipeline stages. Meanwhile, industry-standard Advanced Extendable Interface (AXI) buses are adopted for encapsulating standardized IP cores and building high-speed data exchange bridges between units within Zynq-7000. The developed system processes images of size 32 × 64 in real-time and can generate feature descriptors at a clock rate of 62MHz. Moreover, the method yields a shape feature vector that is computationally light, scalable and rotation invariant. The hardware design is validated using MATLAB for comparative studies

show abstract

“…At present, 3D vision servoing with six degrees of freedom (DOF) for industrial robots mainly includes monocular vision systems (Xia and Weng, 2017), binocular vision systems (Yan-Xiang et al , 2011), profile 3D laser sensors (Koch et al , 2013) and structured light systems (Park et al , 2009; Li et al , 2007; Foix et al , 2011), among others. In binocular vision, two cameras are used to capture an image of the same scene, or one camera is used to capture the same scene from different points of view.…”

Section: Introductionmentioning

confidence: 99%

High-precision six-degree-of-freedom pose measurement and grasping system for large-size object based on binocular vision

Wan

Zhu

et al. 2020

View full text Add to dashboard Cite

Purpose The positioning and grasping of large-size objects have always had problems of low positioning accuracy, slow grasping speed and high application cost compared with ordinary small parts tasks. This paper aims to propose and implement a binocular vision-guided grasping system for large-size object with industrial robot. Design/methodology/approach To guide the industrial robot to grasp the object with high position and pose accuracy, this study measures the pose of the object by extracting and reconstructing three non-collinear feature points on it. To improve the precision and the robustness of the pose measuring, a coarse-to-fine positioning strategy is proposed. First, a coarse but stable feature is chosen to locate the object in the image and provide initial regions for the fine features. Second, three circular holes are chosen to be the fine features whose centers are extracted with a robust ellipse fitting strategy and thus determine the precise pose and position of the object. Findings Experimental results show that the proposed system has achieved high robustness and high positioning accuracy of −1 mm and pose accuracy of −0.5 degree. Originality/value It is a high accuracy method that can be used for industrial robot vision-guided and grasp location.

show abstract

Workpieces sorting system based on industrial robot of machine vision

Cited by 21 publications

References 2 publications

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Design and implementation of shape-based feature extraction engine for vision systems using Zynq SoC

High-precision six-degree-of-freedom pose measurement and grasping system for large-size object based on binocular vision

Contact Info

Product

Resources

About